Pulmonary embolism (PE) is a life-threatening medical condition that occurs when a blood clot travels to the lungs, causing a blockage in one or more pulmonary arteries. Symptoms often include shortness of breath, chest pain, and rapid heart rate, but these can vary in severity and are sometimes difficult to diagnose.
The gold standard for diagnosing PE is CT pulmonary angiography (CTPA), a specialized imaging technique. However, this process can be time-consuming and labor-intensive, often requiring a radiologist to scan through hundreds of CT images to identify blockages manually. Time is of the essence when diagnosing and treating PE; hence, there is a pressing need for a faster and more efficient diagnostic method.
About Pulmonary Embolism CT Dataset
The RSNA STR Pulmonary Embolism Detection Dataset is an expansive and comprehensive collection designed to aid in the machine learning-assisted diagnosis of PE. This dataset is colossal, weighing 980.24 GB and encompassing 1,937,450 files. These files are primarily DICOM (Digital Imaging and Communications in Medicine) images, the standard format for storing and transmitting medical images. The dataset is partitioned into a training set, a public set, and a private set. The training set contains 7279 studies, the public set 650, and the private set 1517.